Clinical Measurement of Range of Motion: Review of
Goniometry Emphasizing Reliability and Validity
Richard L Gajdosik and Richard W Bohannon
PHYS THER. 1987; 67:1867-1872.
The online version of this article, along with updated information and services, can
be found online at: http://ptjournal.apta.org/content/67/12/1867
Collections
This article, along with others on similar topics, appears
in the following collection(s):
Tests and Measurements
e-Letters
To submit an e-Letter on this article, click here or click on
"Submit a response" in the right-hand menu under
"Responses" in the online version of this article.
E-mail alerts
Sign up here to receive free e-mail alerts
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016
Clinical Measurement of Range of Motion
Review of Goniometry Emphasizing Reliability and Validity
RICHARD L. GAJDOSIK
and RICHARD W. BOHANNON
Clinical measurement of range of motion is a fundamental evaluation procedure
with ubiquitous application in physical therapy. Objective measurements of ROM
and correct interpretation of the measurement results can have a substantial
impact on the development of the scientific basis of therapeutic interventions.
The purpose of this article is to review the related literature on the reliability and
validity of goniometric measurements of the extremities. Special emphasis is
placed on how the reliability of goniometry is influenced by instrumentation and
procedures, differences among joint actions and body regions, passive versus
active measurements, intratester versus intertester measurements, and different
patient types. Our discussion of validity encourages objective interpretation of
the meaning of ROM measurements in light of the purposes and the limitations
of goniometry. We conclude that clinicians should adopt standardized methods
of testing and should interpret and report goniometric results as ROM measurements only, not as measurements of factors that may affect ROM.
Key Words: Physical therapy; Tests and measurements, general.
Goniometric measurements are used by physical therapists
to quantify baseline limitations of motion, decide on appropriate therapeutic interventions, and document the effectiveness of these interventions. Probably our most widely used
evaluation procedure, goniometry, can be considered a fundamental part of the "basic science" of physical therapy.
Historically, goniometry developed over the last 60 years
in conjunction with the rapid growth of the field of physical
medicine and rehabilitation. A recent article by Smith provides an interesting account of this development.1 Our review
of the literature revealed a surprising number of articles
describing different methods of measuring ROM, from simple
visual estimation to high-speed cinematography. To most
physical therapists, however, the universal goniometer (ie,
full-circle manual goniometer) remains the most versatile and
widely used instrument in clinical practice. The design of the
universal goniometer and the procedures for its use have been
described in detail in numerous publications.2-9 The articles
by Moore2,3 are particularly comprehensive. Although the
procedures described by the American Academy of Orthopedic Surgeons6 more than 20 years ago are cited frequently in the literature, we encourage the reader to see the
recent publication by Norkin and White8 for current and
complete descriptions.
We believe that physical therapists should master the objective measurement of range of motion and learn to interpret
measurement results correctly. By doing so, the potential
contribution to goniometry to the development of a scientific
basis of physical therapy will be enhanced. The purpose of
Mr. Gajdosik is Associate Professor, Physical Therapy Program, University
of Montana, Missoula, MT, and a doctoral candidate, Department of Anatomy,
School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill,
NC. Direct all correspondence to Physical Therapy Program, University of
Montana, Missoula, MT 59812-1201 (USA).
Mr. Bohannon is Associate Professor, Program in Physical Therapy, University of Connecticut, School of Allied Health Professions, PO Box U-101, 358
Mansfield Rd, Storrs, CT 06268. He was Chief, Department of Physical
Therapy, Southeastern Regional Rehabilitation Center, Cape Fear Valley Medical Center, PO Box 2000, Fayetteville, NC 28302, when this article was written.
this article is to discuss what we view as the two most
important factors affecting objective goniometric measurements, reliability and validity. The related literature will be
reviewed and its implications discussed. The need for additional research to clarify our current knowledge about the
reliability and validity of goniometry will be presented. Because a test must be reliable to be valid,10 we will address the
reliability of ROM measurements before considering their
validity. For the sake of clarity, we will discuss only the
reliability and validity of ROM measurements of the extremities. The basic principles presented, however, can be applied
to ROM measurements of the trunk. This article should help
the reader to gain both a conceptual understanding and a
practical appreciation of clinical ROM measurements. Furthermore, we hope that physical therapists will be stimulated
to develop further the clinical application of goniometry.
RELIABILITY
Reliability in goniometry simply means the consistency or
the repeatability of the ROM measurements, that is, whether
the application of the instrument and the procedures produce
the same measurements consistently under the same conditions. We wish to emphasize from the outset that because few
investigators have used the same study design, comparisons
of the results of reliability studies often are difficult. For
example, studies in which repeated tests are separated by short
time intervals (ie, one hour) may yield very different results
than studies in which repeated tests are separated by longer
time intervals (ie, days or weeks). The reliability of the measurements expresses their reproducibility or stability only in
relation to the time intervals reported. The most accurate
evaluation of the reliability of the instrument and procedures
is determined when short time intervals separate tests, the
classic "test-retest" study design. These results probably will
be more reliable than the results of studies with long time
intervals between tests because the accuracy of the measurements is increased with few uncontrolled variables. Even
though test-retest studies conducted over short time intervals
Volume 67 / Number 12, December 1987
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016
1867
best reflect the reliability of the test, reliability studies conducted over days, or over weeks, are important clinically for
assessing the stability of the measurement so that comparisons
can be made to evaluate patient improvements. Because so
many different study designs and indexes for analysis of
reliability have been reported, we will defer discussion of this
information in this article. We will discuss reliability, but we
refer the reader to the recent article by Stratford et al11 for a
review of specific methodology issues. The reader also may
find the guidelines outlined by Miller9 helpful for establishing
reliability in a clinical setting.
Given that estimating ROM by visual inspection is unreliable when precision and accuracy are needed12 and that
measurements with the goniometer are more reliable estimates of ROM,13 we will consider the reliability of the goniometer and procedures for its use. Although we acknowledge
that the reliability of goniometry is affected by many factors,
we will discuss only the complexity of the movement being
measured, variations among the body regions measured,
whether the movement is active or passive, and whether the
measurement is conducted by the same examiner (intratester
reliability) or by different examiners (intertester reliability).
We believe that the reliability of ROM measurements for
patients may vary with their clinical problems and may differ
from that reported for healthy subjects. Accordingly, we also
will address the potential differences in reliability among
patient types. These differences have important implications
for documenting the effects of physical therapy on ROM.
Instrumentation and Procedures
The comprehensive articles by Moore published in 1949
set the stage for our understanding of the importance of
objective instrumentation and standardized clinical procedures for measuring ROM.2,3 Since that time, most reports
have acknowledged that if the goniometer is reliable, then the
reliability of ROM measurements depends primarily on the
standardization of procedures. Hellebrandt et al found the
universal goniometer to be a "more dependable tool than
devices designed to measure particular motions in specific
joints."12(p307) Salter assumed that inaccurate measurements
were mainly the result of faulty application and concluded
that goniometer error is negligible.4 Hamilton and Lachenbruch found no difference in the accuracy and reliability of
three different goniometers (dorsal, universal, and pendulum)
for measuring finger joint angles when the procedures were
standardized.14 They suggested that the application of the
goniometers should depend on the ease of operation and on
the adaptability of the instrument to the particular clinical
problem (edema, joint enlargements, joint deformities). Using
a Leighton flexometer (a contact pendulum goniometer),
Ekstrand et al compared a method used in clinical practice to
measure passive ROM of the lower extremities (Series A) with
an "optimal" measurement method (Series B).15 In contrast
with the method of Series A, the method of Series B included
covering the examining table with a wood board, standardizing the height of the table at each session, marking bony
landmarks, and checking for faulty movement of the pendulum dial. They found that the mean intratester coefficient of
variation (CV) for Series B measurements (1.9% ±0.7%) was
significantly lower than the mean CV for Series A measurements (7.5% ± 2.9%) and concluded that accurate reproducibility depends on careful measurement technique. Rothstein
et al examined the reliability of three different goniometers
(large metal, large plastic, and small plastic) for measuring
passive elbow and knee positions and reported high interdevice reliability (intraclass correlation coefficient [ICC] > .91)
for the three instruments.16 In a more recent study, Fish and
Wingate reported that improper alignment of the goniometer,
misidentification of bony landmarks, and variations in manual force, all contributed to goniometric error at the elbow.17
Their results indicated that even relatively inexperienced examiners should be able to use goniometers accurately to
measure elbow positions if a standardized method is followed
carefully. The results of these studies clearly demonstrate the
importance of standardizing testing procedures and support
the suggestion by Salter that although error is inherent in the
construction of goniometers, "inaccuracies occur during their
use, mainly from their faulty application. "4(p460)
The reliability of ROM measurements also may be influenced by changes in ROM that result from repeated testing
trials. Cobe, in 1928, studied the active ROM at the wrist and
recommended using the average of several measurements
because individual fluctuations increased the variability of the
measurements.18 Low reported that the average of several
measurements of elbowflexionand wrist extension was more
reliable than one measurement.13 Boone et al, however, studied the reliability of measuring the active ROM of six joint
motions of the upper and lower limbs and concluded that
"one measurement per measurement session is as reliable as
taking the average of repeated measurements in one session."19(p1360) Rothstein et al assessed the reliability of goniometric measurements for passive elbow and knee positions
and found only slight improvement in the intertester reliability by taking the mean score of multiple measurements.16
These conflicting reports may be explained at least partially
by inherent differences in the actions measured (simple or
complex) or by the influence of active versus passive measurements. Atha and Wheatley, for example, studied the effect
of repeated trials of passive straight leg raising (SLR) and
reported a systematic increase in the angle of SLR over the
first four to five trials when measured on the same day.20 The
SLR angle also increased over two consecutive days. Bohannon passively loaded SLR and found systematic increases in
the SLR angle over a three-day period.21 These results indicate
that recording the average of several measurements may
increase the reliability of some ROM measurements, such as
those showing wide variations or systematic increases, although some ROM measurements may be very stable and
therefore reliable without averaging. Physical therapists
should understand, however, that posttreatment increases in
some measurements may be a normal function of increased
compliance of tissues from repeated testing over time. More
research is needed to examine the effects of repeated measurements, particularly in relation to the specific actions measured and the methods used.
Differences Among Joint Actions and Among
Structure and Function of Body Regions
The reliability of measuring ROM of the extremities is
affected by the complexity of the actions measured and by
the inherent structural and functional differences of the action. Hellebrandt et al found that wrist flexion, medial rotation of the shoulder, and shoulder abduction "did not lend
themselves to reliable repetitive measurement. "12(p306) Low
also reported that the reliability of goniometric measurements
varies in different joints.13 Boone et al reported that the
intertester reliability was higher for upper extremity motions
(r = .86) than for lower extremity motions (r = .58).19 These
PHYSICAL THERAPY
1868
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016
results suggest that reliability of measuring ROM is specific
to the action measured and to regional structure and function.
For example, measurements of the elbow, generally considered a simple hinge joint, show less day-to-day variation in
ROM than measurements of the wrist,13 the movement of
which is affected by multiple joints and numerous muscles
crossing these joints. Although the ROM measurements of
movements that are influenced by movement of adjacent
joints or by multiple joint muscles might be expected to be
less reliable than the ROM measurements of simple hinge
joints, strict standardization of procedures should increase
reliability. Gajdosik and Lusin22 and Gajdosik et al23 have
demonstrated that even complex movements can be measured
reliably when the measurement procedures are controlled
strictly. The importance of applying standardized procedures
was emphasized in these reports. Physical therapists should
realize, however, that measuring complex actions reliably may
be more difficult than measuring simple actions because of
the greater potential for error resulting from large fluctuations
within the same individual13,18,24 or because of systematic
changes over time.20,21
Passive Versus Active Measurements
Although few studies compared directly the reliability of
measuring active ROM with that of measuring passive ROM,
sufficient evidence is available to suggest that passive ROM is
more difficult to measure reliably than active ROM. Amis
and Miller acknowledged this problem as follows: "Passive
movements are extremely difficult to reproduce, because the
stretching of soft tissues at the limits of motion depends on
the force applied to the limb, which must, therefore, be
carefully controlled. "25(p578) Wagner examined pronation and
supination of the forearm and found that the variability of
passive flexibility generally was higher than the variability of
active flexibility.26 The range of passive forearm movement
was nonlinearly dependent on the external torque applied,
and the variability of the passive movement increased with
decreasing external torque. Bird and Stowe reported that the
error for measurements of passive movements of the wrist
was greater than the error for measurements of active movements.27 Pandya et al studied the reliability of passive ROM
measurements in seven upper and lower extremity movements in children with Duchenne muscular dystrophy and
reported good intratester reliability (ICC range, .81-.94) for
all movements.28 Intertester reliability, however, was lower
than intratester reliability for all measurements (ICC range,
.25-.91). The lowest intertester reliability value was for measurements of iliotibial band contractures. The authors hypothesized that because the position and the procedure were
standardized, "the force exerted by the therapist during the
passive movement may be the variable that caused the goniometric discrepancy."28(p1341) These results indicate that obtaining high reliability among different examiners for measurements of complex passive movements is more difficult
than for measurements of simple passive movements.
The value of both passive and active ROM measurements
cannot be disputed, and we acknowledge their importance,
particularly with different patient problems. Therapists, however, should consider the potential for greater variations and
decreased reliability that may occur with passive measurements and attempts to overcome this problem. Using standardized testing procedures among physical therapists in a busy
clinic should be given special consideration. Perhaps force
dynamometers29 (including those that are hand-held) could
be used to standardize the amount of passive force applied
and thereby decrease the potential for error. We recommend
clinical studies to explore these possibilities.
Intratester Versus Intertester Measurements
In 1949, Hellebrandt et al studied the reliability of active
movements of the upper extremity.12 Even though the results
of their study may have been influenced by the broad variety
of patients examined, they did show that a well-trained therapist can measure ROM with a high degree of reliability. The
authors stated, however, the following: "Where the interreliability of different physical therapists has not been established, different observers should not be used interchangeably
to obtain measurements on the same patient."12(p307)
Since this study by Hellebrandt et al, many investigations
have demonstrated in healthy subjects that intratester reliability is higher than intertester reliability.13,14,19,30-33 Hamilton
and Lachenbruch reported greater intertester variance than
intratester variance for measuring finger joint angles.14 Low
measured full elbow flexion and full wrist flexion and showed
that the error between observers was much larger than the
error within the same observer.13 In a well-documented study
that is cited frequently in the literature on goniometry, Boone
et al reported that the intertester reliability was lower than the
intratester reliability for weekly measurements of active ROM
of the upper and lower extremities.19 As stated earlier in this
article, they found higher intertester reliability for upper extremity measurements than for lower extremity measurements. Based on their results, they suggested that when more
than one tester measures the same motion, changes in ROM
should exceed 5 degrees in upper extremity measurements
and 6 degrees in lower extremity measurements to document
"improvement." In light of the difficulty of measuring the
ROMs of different actions with similar reliability, we believe
that these guidelines should be reconsidered. These guidelines
may be particularly difficult to apply to different patient types.
Even studies of healthy subjects have demonstrated different
outcomes. Ellis and Stowe found 1% to 5% error for intratester measurements, but as much as 5% to 10% error for
intertester measurements of the ROM of the hip.30 Thus, if
hip flexion ROM was about 130 degrees, then 5% to 10%
error would equal about 7 degrees to 13 degrees. Solgaard et
al reported that intratester CVs were as much as 5% to 8%
and that intertester CVs were 6% to 10% for ROM measurements of the wrist.31 We encourage physical therapists to be
cautious accepting generalizations about the reliability of
ROM measurements, such as those suggested by Boone et
al,19 in light of our current knowledge about the potential for
variations among different ROM measurements.
Smith and Walker demonstrated higher intratester reliability (r = .90) than intertester reliability (r = .70) for knee and
elbow ROM in healthy elderly persons.32 In a later study,
Walker et al reported greater variability between testers than
within testers for active ROM measurements of the upper and
lower extremities of elderly subjects.33
Generally, the results of intratester and intertester reliability
studies with patients support the findings reported for healthy
subjects. Using patients as subjects, Rothstein et al found that
intratester reliability for flexion and extension of the knee and
elbow joints was high and that intertester reliability, although
lower, also was high, except for knee extension (r range, .57.80).16 Pandya et al studied seven upper and lower extremity
Volume 67 / Number 12, December 1987
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016
1869
joint limitations in children with Duchenne muscular dystrophy and reported that intratester reliability was high (ICC
range, .81 -.94), but intertester reliability was lower with a
wide variation (ICC range, .25-.91).28
Recent evidence suggests that the reliability of ROM measurements may be influenced directly by the type of patient
problem. Ashton et al examined the day-to-day reliability of
measuring ROM of the hip in children with spastic diplegia
and found that the reliability usually was lower for children
with a moderate condition than for children with a mild
condition. Because the level of reliability was low, they stated
that ROM measurements "are not sufficiently reliable to be
used in studies of cerebral palsy. "34(p91) Harris et al examined
further the goniometric reliability for a child with cerebral
palsy (ie, spastic quadriplegia).35 They found that intratester
reliability exceeded intertester reliability and, as expected, that
a wide variation existed in measurements, both within examiners and among examiners. The authors concluded that
a difference of ± 10 to 15 degrees in ROM over time does
not signify a meaningful change in a child with severe spastic
cerebral palsy. Bartlett et al compared four methods of measuring hip extension (prone extension, Thomas test, Mundale,
and pelvifemoral angle) in patients with spastic diplegia and
meningomyelocele.36 They found that the Thomas test was
particularly difficult to apply in patients with spastic diplegia
and that improved reliability most likely will result in this
patient group by using one of the other techniques. They also
reported that the least reliable technique in the meningomyelocele group was that of Mundale, probably because of difficulties associated with identifying bony landmarks in the
presence of obesity and deformities. These researchers also
found intratester reliability to be superior to intertester reliability. The results of these few studies with patients clearly
show that the reliability of ROM measurements is influenced
by the specific patient problem, and the results suggest a
tremendous need for studies of the reliability of measuring
ROM among the different patient types within the specialized
areas of physical therapy of orthopedics and neurology.
VALIDITY
Currier has stated that the validity of a measurement "constitutes the degree to which an instrument measures what it
is purported to measure; the extent to which it fulfills its
purpose."10(p166) In goniometry, we must be confident that the
goniometer and measurement procedures are accurate and
that we understand the meaning of the measurement results.
To do so, we must show that the measurement procedures
are consistent with our interpretation of the results. Do we
understand the purpose of goniometry and the limitations of
goniometry? Are we measuring what we say we are measuring?
The primary purpose of goniometry is to measure ROM of
the musculoskeletal system of the human body. To fulfill this
purpose, the goniometer was designed as a modification of
the protractor, a device known to represent accurately the
degree intervals of a circle. If the accuracy and, therefore, the
validity of goniometers are in question, the degree units can
be compared simultaneously against known angles.32,33 A
strong relationship establishes concurrent validity, a type of
criterion-related validity. This comparison can detect any
systematic errors resulting from faulty construction so that
proper adjustments can be made. Although small errors in
the construction of goniometers may exist, the instruments
generally are accepted as valid clinical tools. They do, however, have limitations. For example, the measurements ob-
tained are limited to degree units of a circle and, as a result,
we accept that the movements we measure have fixed axes of
motion about which movement occurs. Smidt, however, demonstrated that this assumption is not true.37 Because of other
motions within joints, such as articular sliding and rotation,
the axes of motion are not fixed. In a strict sense, the validity
of representing movement of body parts by units of a circle
can be challenged. Nevertheless, we accept this limitation and
are confident that the ROM measured closely approximates
movement around a central point, that is, that the ROM
measurements are clinically valid.
Another limitation of goniometric measurements is that
they are recorded in degree units and, therefore, only ROM
can be measured. This limitation is not a problem if the
results are interpreted within this restriction, but physical
therapists inadvertently may reach beyond this restriction and
invalidly expand the meaning of the results. Many factors
may affect the outcome of ROM measurements—edema,
pain, adhesions, strength deficits, and muscle hypertrophy—
but ROM measurements are never measures of factors other
than ROM. The validity of ROM measurements is very
specific. In the extremities, ROM is recorded in degrees,
whereas the factors that may affect ROM must be measured
by different methods with equally different measurement
units. This point is extremely important and worthy of
discussion.
A classic example of how physical therapists may expand
the meaning of ROM measurements and invalidly interpret
the results can be found by examining the SLR test. Most
physical therapists are aware that the maximal angle of SLR
in relation to the horizontal plane has been used historically
to provide information about hamstring muscle length (HML)
and hamstring muscle flexibility (HMF). But can we really
use this test to measure HML and HMF? The validity of
using the SLR test to assess HML and HMF has been challenged recently, and the results of studies are giving new
meaning and validity to the test. Some study results have
indicated that during SLR the pelvis moves in conjunction
with the lower limb38-40; thus, the origin of the hamstring
muscles is not fixed. The reports also indicated that SLR may
be confounded by pain from nerve root irritation39,40 and by
the position of the ankle and stretching structures other than
the hamstring muscles (eg, sciatic nerve, enveloping deep
fascia of the lower limb).40 These findings do not mean that
the SLR test is invalid for measuring the maximal angle of
SLR in relation to the horizontal plane, but that these results
are limited to this interpretation alone. Hamstring muscle
length may affect the maximal angle of SLR, but the SLR
test does not measure HML directly. If the results of SLR
were compared with the results of linear measurements of
HML and if the two measurements were strongly related,
then we could say with confidence that the angle of SLR
indirectly represents HML—concurrent validity would be
demonstrated.
We are unaware of published studies demonstrating a strong
relationship between the angle of SLR and HML. Tardieu et
al showed that the tibeocalcaneal angle corresponds to the
length of the triceps surae muscle group in subjects of the
same height41; therefore, the angles of some joints may be
valid, indirect indications of muscle length. Obviously, more
research is needed to validate claims that the maximal angle
of SLR represents HML.
The SLR test also has been widely accepted as a measure
of HMF. Individuals with a small SLR angle are considered
PHYSICAL THERAPY
1870
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016
to have poor HMF, and those with a large SLR angle are
believed to have good HMF. As with using the SLR test to
measure HML, using the test to indicate theflexibilityof the
muscles may be incorrect and, therefore, invalid. Even though
the maximal ROM of a joint or combined motions of multiple
joints have been interpreted historically to yield information
about the flexibility of the motions examined, no scientific
basis exists for these conclusions. Passiveflexibility,or exten­
sibility, of skeletal muscles is a physiological property and is
defined by the length-tension relationship of the tissues. This
relationship can be represented as "stiffness"—the ratio of the
change in passive muscle tension (ΔP) to the change in muscle
length (ΔL) (ΔP/ΔL)—or as "compliance"—the ratio of the
change in muscle length (ΔL) to the change in passive muscle
tension (ΔP) (ΔL/ΔP).42,43 Whether the maximal angle of SLR
represents the length-tension relationship of the hamstring
muscles is unknown and currently is being investigated by
the first author. The same physiological principles apply to
ROM measurements throughout the body, and studies are
needed to examine the validity of using ROM measurements
to represent flexibility.
As we discussed earlier in this article, the standardization
of testing procedures improves reliability, and goniometric
measurements must be reliable to be valid. Reliable measure­
ments, however, do not ensure that the measurements are
valid; invalid measurements can be measured reliably. One
obvious example already discussed would be to accept reliable
measurements of the angle of SLR as valid measurements of
HML or HMF when no evidence exists to indicate that this
conclusion is true. This problem results primarily from inter­
preting the results incorrectly in relation to the measurement
procedures. The content validity of the test (ie, how well the
test reflects what is being measured), which basically is judg­
mental, is not established. As we shall see shortly, accurate
judgment of the content validity of ROM measurements may
vary depending on the complexity of the test. Again, we must
ask ourselves, Are we measuring what we say we are
measuring?
Physical therapists judge the validity of most ROM meas­
urements based on their anatomical knowledge and their
applied skills of visual inspection, palpation of bony land­
marks, and accurate alignment of the goniometer. Generally,
the accurate application of knowledge and skills, combined
with interpreting the results as measurements of ROM only,
provide sufficient evidence to ensure content validity. This
finding is particularly true with simple measurements, such
as those of the elbow and knee. The reliability and the validity
of even simple measurements, however, may be decreased
because of patient differences that we cannot control. Obesity
and variations in bony structures can make accurate visual
inspection and bony palpation very difficult. Perhaps one
reason Boone et al19 found that the reliability of lower extrem­
ity ROM measurements was less than the reliability of upper
extremity ROM measurements was because of greater diffi­
culty accurately locating bony landmarks and aligning the
goniometer with the landmarks of the lower extremity. Fac­
tors such as the size and the weight of the extremity measured
also will affect the clinician's ability to manipulate the part
and thus the reliability and the validity of the measurements.
Even though physical therapists may have confidence in
the validity of ROM measurements based on judgment, ap­
plying more objective methods of motion analyses to study
ROM can strengthen our knowledge of the content validity
by demonstrating criterion-related validity. Comparing the
results of goniometric measurements with those taken from
photographs is one method that can be used to demonstrate
validity. Fish and Wingate assessed the accuracy and sources
of error in goniometry at the elbow using still photography as
a reference standard.17 If still photography is used, the pho­
tographic procedures must be standardized to ensure their
validity. Variations in the position of the camera in relation
to the subject and the type of lens used (ie, 35 mm vs 50 mm)
are only two of many potential sources of error. Still photog­
raphy also requires marking anatomical landmarks with skin
marks, and movement of the bones under the skin may cause
additional error.
Cinematography also has been used to study the validity of
motions routinely measured by goniometry.38-40,44 The results
of these studies have provided new insights into the content
validity of ROM measurements of the hip. The implications
of the results of studies of the SLR test have been discussed.
In addition to these results, when active and passive unilateral
and bilateral hip flexion with knee flexion was examined by
cinematography,44 the displacement of skin marks placed over
bony landmarks indicated that the hip flexion movement is
composed of two movements: 1) flexion of the thigh on the
pelvis and 2) pelvic rotation. The implication of these results
is that the traditional method of measuring hipflexionprob­
ably includes movement of both the femur and the pelvis,
just as shoulder abduction incorporates movement of both
the humerus and the scapula.45,46 Consequently, our knowl­
edge of ROM measurements of the hip flexion movement
takes on new and more valid meaning.
The application of cinematography and other forms of
motion analysis (eg, electromyography, electrogoniometry,
videotape recording) have the potential to add substantially
to our understanding of ROM measurements. These tech­
niques may be particularly helpful for studying the relation­
ship of goniometric measurements to the functional ROM
needed for patients to perform various activities of daily living.
As with still photography, cinematography is limited by sim­
ilar procedural problems, such as the positions of the camera
and the subject and documenting movement with skin marks
over bony landmarks that may move under the skin. Even
so, careful application of these techniques can contribute to
our knowledge of ROM measurements.
The most powerful method by which the validity of ROM
measurements can be studied is radiography. Lawrence con­
firmed the validity of goniometry at the knee by comparing
goniometric measurements with measurements from radio­
graphs of cadavers and patients.47 More recently, however,
Enwemeka compared goniometric knee measurements with
radiographic bone angle measurements in healthy adults and
reported that "within the first 15 degrees of knee flexion,
goniometric measurement of joint excursion may be remark­
ably wrong. "48(p47) He indicated that the error resulted from
knee rotation, which changes the relative position of the bony
landmarks used in goniometry.
In a classic study, Freedman and Munro demonstrated
through radiographs that shoulder abduction results from the
relative contributions of both scapular movements and glenohumeral movements.45 The results of this study were repro­
duced by Doody and Waterland using special goniometric
procedures.46 The results of these two studies confirmed that
measurements of total shoulder abduction are valid only for
the combined movements of the scapula on the thorax and
the humerus on the scapula. As a result, physical therapists
consider these movements when measuring ROM of the
Volume 67 / Number 12, December 1987
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016
1871
shoulder, and valid clinical procedures have been developed
to isolate the two component motions.8(pp30-31),49 Objective
methods of motion analysis other than radiography also have
contributed to our knowledge of the complex movements that
occur at the shoulder. Blakely and Palmer recently showed
that medial rotation of the humerus, not lateral rotation as
generally believed, accompanies active and passive shoulder
flexion.50 This finding was reported first for healthy men and
women,50 and subsequently confirmed using a fresh cadaver.51
The therapeutic implications then were studied by analyzing
the flexion-abduction-lateral rotation pattern of proprioceptive neuromuscular facilitation.52 We commend the authors
for their systematic and thorough methods of inquiry and
encourage similar endeavors. As the body of knowledge of
physical therapy and movement analysis advances, we expect
that findings similar to these will continue to change our
understanding of movement and the validity of ROM
measurements.
CONCLUSIONS
For most applications to the extremities, the universal fullcircle goniometer probably is the preferred instrument for
measuring ROM. Although the literature abounds with studies about the reliability of ROM measurements, applying their
results to clinical practice may be difficult because of differences in study designs. Because the reliability of goniometry
is dependent on a host of factors, such as differences among
the motions measured, methods of application, and variations
among different patient types, clinicians working in the same
setting should adopt standardized methods of testing. These
methods would require a thorough orientation for both students and staff. Physical therapists should be careful in the
interpretation and reporting the results of goniometric findings. As a rule, ROM measurements are just that, not measurements of muscle "tightness," the length of specific structures, or other factors that may affect ROM. We recommend
continued research so that physical therapists may gain a
complete and objective understanding of the meaning of
goniometric measurements.
REFERENCES
.1. Smith DS: Measurement of joint range: An overview. Clin Rheum Dis
8:523-531,1982
2. Moore ML: The measurement of joint motion: Part I. Introductory review
of the literature. Phys Ther Rev 29:195-205, 1949
3. Moore ML: The measurement of joint motion: Part II. The technic of
goniometry. Phys Ther Rev 29:256-264, 1949
4. Salter N: Methods of measurement of muscle and joint functions. J Bone
Joint Surg [Br] 37:474-491, 1955
5. Defibaugh JJ: Measurement of head motion: Part I. A review of methods
of measuring joint motion. Phys Ther 44:157-163, 1964
6. Joint Motion: Method of Measuring and Recording. Chicago, IL, American
Academy of Orthopedic Surgeons, 1965
7. Lusin GF, Gajdosik RL, Miller KE: Goniometry: A review of the literature.
Athletic Training 14(3):161-164, 1979
8. Norkin CC, White DJ: Measurement of Joint Motion: A Guide to Goniometry. Philadelphia, PA, F A Davis Co, 1985
9. Miller PJ: Assessment of joint motion. In Rothstein JM (ed): Measurement
in Physical Therapy: Clinics in Physical Therapy. New York, NY, Churchill
Livingstone Inc. 1985, vol 7, pp 103-136
10. Currier DP: Elements of Research in Physical Therapy, ed 2. Baltimore,
MD, Williams & Wilkins, 1984, pp 166-167
11. Stratford P, Agostino V, Brazeau C, et al: Reliability of joint angle measurements: A discussion of methodology issues. Physiotherapy Canada
36:5-9, 1984
12. Hellebrandt FA, Duvall EN, Moore ML: The measurement of joint motion:
Part III. Reliability of goniometry. Phys Ther Rev 29:302-307, 1949
13. Low JL: The reliability of joint measurement. Physiotherapy 62:227-229,
1976
14. Hamilton GF, Lachenbruch PA: Reliability of goniometers in assessing
finger joint angle. Phys Ther 49:465-469, 1969
15. Ekstrand J, Wiktorsson M, Oberg B, et al: Lower extremity goniometric
measurements: A study to determine their reliability. Arch Phys Med
Rehabil 63:171-175, 1982
16. Rothstein JM, Miller PJ, Roettger RF: Goniometric reliability in a clinical
setting: Elbow and knee measurements. Phys Ther 63:1611-1615, 1983
17. Fish DR, Wingate L: Sources of goniometric error at the elbow. Phys Ther
65:1666-1670,1985
18. Cobe HM: The range of active motion at the wrist of white adults. J Bone
Joint Surg 10:763-774, 1928
19. Boone DC, Azen SP, Lin CM, et al: Reliability of goniometric measurements. Phys Ther 58:1355-1360, 1978
20. Atha J, Wheatley DW: The mobilising effects of repeated measurements
on hip flexion. Br J Sports Med 10:22-25, 1976
21. Bohannon RW: Effect of repeated eight-minute muscle loading on the
angle of straight-leg raising. Phys Ther 64:491-497, 1984
22. Gajdosik RL, Lusin GF: Hamstring muscle tightness: Reliability of an activeknee-extension test. Phys Ther 63:1085-1088, 1983
23. Gajdosik RL, Simpson R, Smith R, et al: Pelvic tilt: Intratester reliability of
measuring the standing position and range of motion. Phys Ther 65:169174,1985
24. Hewitt D: The range of active motion at the wrist of women. J Bone Joint
Surg 10:775-787, 1928
25. Amis AA, Miller JH: The elbow. Clin Rheum Dis 8:571-593, 1982
26. Wagner C: Determination of the rotary flexibility of the elbow joint. Eur J
Appl Physiol 37:47-59, 1977
27. Bird HA, Stowe J: The wrist. Clin Rheum Dis 8:559-569, 1982
28. Pandya S, Florence JM, King WM, et al: Reliability of goniometric measurements in patients with Duchenne muscular dystrophy. Phys Ther
65:1339-1342,1985
29. Bohannon RW, Lieber C: Cybex® II isokinetic dynamometer for passive
load application and measurement: Suggestion from the field. Phys Ther
66:1407,1986
30. Ellis Ml, Stowe J: The hip. Clin Rheum Dis 8:655-675, 1982
31. Solgaard S, Carisen A, Kramhoft M, et al: Reproducibility of goniometry of
the wrist. Scand J Rehabil Med 18:5-7, 1986
32. Smith JR, Walker JM: Knee and elbow range of motion in healthy older
individuals. Physical and Occupational Therapy in Geriatrics 2(4):31-38,
1983
33. Walker JM, Sue D, Miles-Elkousy N, et al: Active mobility of the extremities
in older subjects. Phys Ther 64:919-923, 1984
34. Ashton BB, Pickles B, Roll JW: Reliability of goniometric measurements of
hip motion in spastic cerebral palsy. Dev Med Child Neurol 20:87-94, 1978
35. Harris SR, Smith LH, Krukowski L: Goniometric reliability for a child with
spastic quadriplegia. J Pediatr Orthop 5:348-351, 1985
36. Bartlett MD, Wolf LS, Shurtleff DB, et al: Hip flexion contractures: A
comparison of measurement methods. Arch Phys Med Rehabil 66:620625, 1985
37. Smidt GL: Biomechanical analysis of knee flexion and extension. J Biomech
6:79-92, 1973
38. Bohannon RW: Cinematographic analysis of the passive straight-leg-raising test for hamstring muscle length. Phys Ther 62:1269-1274, 1982
39. Bohannon RW, Gajdosik RL, LeVeau BF: Contribution of pelvic and lower
limb motion to increases in the angle of passive straight leg raising. Phys
Ther 65:474-476, 1985
40. Gajdosik RL, LeVeau BF, Bohannon RW: Effects of ankle dorsiflexion on
active and passive unilateral straight leg raising. Phys Ther 65:1478-1482,
1985
41. Tardieu G, Lespargot A, Tardieu C: To what extent is the tibia-calcaneum
angle a reliable measurement of the triceps surae length? RadiologicaL
correction of the torque-angle curve (III). Eur J Appl Physiol 37:163-171,
1977
42. Walker SM: Tension and extensibility changes in muscles suddenly
stretched during tetanus. Am J Physiol 172:37-41, 1953
43. Botelho SY, Cander L, Guiti N: Passive and active tension length diagrams
of intact skeletal muscle in normal women of different ages. J Appl Physiol
7:93-95, 1954
44. Bohannon RW, Gajdosik RL, LeVeau BF: Relationship of pelvic and thigh
motions during unilateral and bilateral hip flexion. Phys Ther 65:15011504,1985
45. Freedman L, Munro RR: Abduction of the arm in the scapular plane:
Scapular and glenohumeral movements—A roentgenographic study. J
Bone Joint Surg [Am] 48:1503-1510, 1966
46. Doody SG, Waterland JC: Shoulder movements during abduction in the
scapular plane. Arch Phys Med Rehabil 51:595-604, 1970
47. Lawrence MR, referred to in Johnson F: The knee. Clin Rheum Dis 8:677701,1982
48. Enwemeka CS: Radiographic verification of knee goniometry. Scand J
Rehabil Med 18:47-49, 1986
49. Hoppenfeld S: Physical Examination of the Spine and Extremities. New
York, NY, Appleton-Century-Crofts, 1976, p 923
50. Blakely RL, Palmer ML: Analysis of rotation accompanying shoulder flexion.
Phys Ther 64:1214-1216, 1984
51. Palmer ML, Blakely RL: Documentation of medial rotation accompanying
shoulder flexion: A case report. Phys Ther 66:55-58, 1986
52. Blakely RL, Palmer ML: Analysis of shoulder rotation accompanying a
proprioceptive neuromuscular facilitation approach. Phys Ther 66:12241227, 1986
1872
PHYSICAL THERAPY
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016
Clinical Measurement of Range of Motion: Review of
Goniometry Emphasizing Reliability and Validity
Richard L Gajdosik and Richard W Bohannon
PHYS THER. 1987; 67:1867-1872.
Cited by
This article has been cited by 39 HighWire-hosted
articles:
http://ptjournal.apta.org/content/67/12/1867#otherarticles
Subscription
Information
http://ptjournal.apta.org/subscriptions/
Permissions and Reprints http://ptjournal.apta.org/site/misc/terms.xhtml
Information for Authors
http://ptjournal.apta.org/site/misc/ifora.xhtml
Downloaded from http://ptjournal.apta.org/ by guest on March 4, 2016